Project Summary Glucocorticoid-induced osteoporosis (GIO) is the most common secondary form of osteoporosis. Glucocorticoid users are at a higher risk for all types of fragility fractures compared to non-users. Effective medications exist that can reduce fracture risk in GIO. However, standard-of-care methods used to diagnose osteoporosis ? dual-energy x-ray absorptiometry (DXA) estimation of areal bone mineral density (BMD) and FRAX ? underestimate fracture risk in glucocorticoid users. As a result, GIO is under-diagnosed and under- treated. The lack of a relationship between BMD and fracture risk in GIO means that glucocorticoids negatively affect bone quality and strength in a way not captured by BMD. Recently, we demonstrated the feasibility of imaging proximal femur microarchitecture in vivo on a clinical 3T MRI scanner. We have shown that such microarchitectural assessment is reproducible and provides information about bone quality that is not captured by DXA. We will now leverage the unique resources of Radiology (novel imaging test of proximal femur microarchitecture) and Rheumatology (large glucocorticoid-using patient population at a lupus center nationally known for clinical care and research) at our institution. In two specific aims, we will determine: 1) the added value of the novel MRI test, beyond DXA, for sensitively monitoring longitudinal, detrimental changes in bone microarchitecture and strength in the proximal femur, a standard site of BMD assessment used in FRAX and a clinically important fracture site and 2) the value of proximal femur microarchitectural assessment, beyond that of DXA/FRAX, for the discrimination of GIO subjects with fracture from controls without fracture. To achieve Aim 1, we will recruit 40 lupus patients newly prescribed glucocorticoids and 40 lupus patients managed without glucocorticoids (both as standard-of-care). We hypothesize that after adjusting for areal BMD and disease severity, glucocorticoid users will demonstrate 12 month detrimental changes in femoral neck cortical thickness, trabecular thickness, separation, number, connectivity, and whole femur stiffness of greater magnitude than in controls. To achieve Aim 2, we will separately recruit 138 long-term (> 12 month), glucocorticoid using lupus patients (expect 41 with fracture, 97 without fracture). We will build multivariate, logistic regression models predictive of fracture status and compare the accuracy of different models (DXA, FRAX, DXA/FRAX plus microarchitectural measures) for prediction of fracture status. We hypothesize that the addition of microarchitectural measures to DXA/FRAX will improve model accuracy for prediction of fracture status compared to DXA/FRAX alone. Successful execution of this project will demonstrate the feasibility of detecting currently occult, skeletally fragile glucocorticoid users who are in need of osteoporosis therapy. By improving clinicians' ability to accurately diagnose GIO, we will improve clinicians' ability to accurately treat GIO or enroll patients in clinical trials for bone-strengthening therapies. This will reduce the burden of glucocorticoid-induced fractures on society.